Data Collectionmedium
0:00.0

A researcher uses the Horvitz-Thompson estimator to estimate the population total Y=i=1NyiY = \sum_{i=1}^N y_i. If each unit ii has an inclusion probability πi\pi_i, the estimator is Y^HT=isyiπi\hat{Y}_{HT} = \sum_{i \in s} \frac{y_i}{\pi_i}. Given a population of size N=5N=5 with units y={10,20,30,40,50}y = \{10, 20, 30, 40, 50\} and a sample of n=2n=2 where units are selected with probabilities πi\pi_i proportional to their values (PPS), what is the expected value E[Y^HT]E[\hat{Y}_{HT}]?